Share Email Print

Proceedings Paper

Inter-comparison of aerosol optical thickness from MODIS, MISR, and OMI using measurements from solar radiation stations in China
Author(s): Maohua Guo; Ling Sun; Xiaofeng Xu
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Aerosol observations are essential for understanding the Earth's radiation budget and the complexities of climate change. East Asia plays a significant part in aerosol loading. Several satellite-based aerosol optical thickness (AOT) products have provided long-term monitoring of aerosol properties, but they differ due to different sensors and algorithms. The discrepancies among them result in the uncertainty of the understanding of the long-time series of aerosols in East Asia, especially in China. China has 17 first-class solar radiation stations providing hourly accumulated solar measurements since 1993. The AOT can be retrieved by a broadband extinction method, and the retrievals have been validated in comparison with AERONET (Aerosol Robotic Network). The AOT measurements from radiation stations provide more ground-based truth in satellite product validation in China. This research compares the AOT of several aerosol operational products, namely the Moderate Resolution Imaging Spectroradiometer (MODIS), the Multiangle Imaging Spectroradiometer (MISR), and the Ozone Monitoring Instrument (OMI), with ground-based measurement from China solar radiation stations from 2002 to 2012. MODIS products from Dark Target (DT) and Deep Blue (DB) algorithms, OMI products from Multi-wavelength (MW) and Near-UV (UV) algorithms and MISR product are evaluated. Analysis shows that (1) for MODIS DT, there are few retrievals in arid/semi-arid regions and winter, about 52.04% pairs fall within the error range (±(0.15τ+0.05)), and RMSE is 0.24; (2) for MODIS DB, more retrievals could be provided in arid/semi-arid regions and winter, data percentage within error range and RMSE are 54.96%, 0.17 in arid/semi-arid regions, and 35.84%, 0.42 in other regions; (3) for MISR, AOT tends to be underestimated when AOT larger than 0.2, the data percentage and RMSE are 83.88%, 0.11 in arid/semi-arid regions, and 50.94%, 0.25 in other regions; (4) for OMI, UV could provide more effective retrievals than MW, MW AOT tends to be overestimated with data percentage and RMSE of 21.79%, 0.28 in arid/semi-arid regions, and 11.87%, 0.63 in other regions; UV AOT tends to be overestimated in arid/semi-arid regions with data percentage and RMSE of 11.70% and 0.30, while underestimated in other regions when AOT larger than 0.3, and the data percentage and RMSE are 40.23% and 0.41. In arid/semi-arid regions, the overall performance in decreasing order is MISR, DB, UV (after system error correction), and MW, while in other regions, the overall performance in decreasing order is DT, MISR, DB, UV and MW. The results are consistent with previous validations based on sunphotometer measurements.

Paper Details

Date Published: 6 September 2016
PDF: 9 pages
Proc. SPIE 9876, Remote Sensing of the Atmosphere, Clouds, and Precipitation VI, 98763F (6 September 2016); doi: 10.1117/12.2228000
Show Author Affiliations
Maohua Guo, National Satellite Ocean Application Service (China)
Ling Sun, National Satellite Meteorological Ctr. (China)
Xiaofeng Xu, Nanjing Univ. of Information Science and Technology (China)

Published in SPIE Proceedings Vol. 9876:
Remote Sensing of the Atmosphere, Clouds, and Precipitation VI
Eastwood Im; Raj Kumar; Song Yang, Editor(s)

© SPIE. Terms of Use
Back to Top